TY - JOUR
T1 - Fast-Tactical Diffusion for On-Board AAV Spectrum-Level Signal Deception
AU - Xu, Lin
AU - Wu, Yihan
AU - Zhang, Zeyu
AU - Feng, Lingyun
AU - Zhu, Chao
AU - Wang, Fangxin
AU - An, Jianping
N1 - Publisher Copyright:
© 2014 IEEE.
PY - 2026/6/1
Y1 - 2026/6/1
N2 - Spectrum deception has found broad utility across multiple domains, including electronic warfare, tactical countermeasures, and adversarial sensing suppression. However, generating complex time-frequency signatures relies on sophisticated signal processing pipelines, which poses significant challenges for AAV and other IoT platforms with severely constrained onboard computational resources. Moreover, the limited onboard capability further restricts rapid signal synthesis and adaptation, failing to meet the strict rapid-response requirements of the battlefield. To address this dilemma, we propose the fast-tactical signal deception framework (FT-SDF), a specialized generative architecture optimized for real-time signal synthesis. We formulate a novel spectrotemporal diffusion dynamics mechanism that innovatively incorporates additional spectral blurring and reverse process variance, jointly optimizing noise prediction and variance, which is necessary to preserve fine-grained spectral structures and key time-frequency signatures across different modulation schemes. Notably, to ensure strict adherence to communication protocols, we introduce a lightweight spectrum-context encoder (LSCE) that uses a dual-domain embedding strategy for physics-aware conditioning. Furthermore, to enable rapid inference, we develop a variance-aware acceleration mechanism that exploits learned spectral uncertainty to guide a dynamic warm-start schedule, thereby drastically compressing the sampling trajectory. Extensive evaluations on a systematically reconstructed RadioML benchmark demonstrate that FT-SDF outperforms state-of-the-art baselines. Specifically, it achieves a 59.3% reduction in sampling iterations (more than twofold inference speedup), while maintaining both high statistical fidelity (Fréchet inception distance (FID) <15) and industrial-grade precision (error vector magnitude (EVM) ≤ 14%), demonstrating the feasibility of rapid, controllable generative AI in complex electromagnetic environments.
AB - Spectrum deception has found broad utility across multiple domains, including electronic warfare, tactical countermeasures, and adversarial sensing suppression. However, generating complex time-frequency signatures relies on sophisticated signal processing pipelines, which poses significant challenges for AAV and other IoT platforms with severely constrained onboard computational resources. Moreover, the limited onboard capability further restricts rapid signal synthesis and adaptation, failing to meet the strict rapid-response requirements of the battlefield. To address this dilemma, we propose the fast-tactical signal deception framework (FT-SDF), a specialized generative architecture optimized for real-time signal synthesis. We formulate a novel spectrotemporal diffusion dynamics mechanism that innovatively incorporates additional spectral blurring and reverse process variance, jointly optimizing noise prediction and variance, which is necessary to preserve fine-grained spectral structures and key time-frequency signatures across different modulation schemes. Notably, to ensure strict adherence to communication protocols, we introduce a lightweight spectrum-context encoder (LSCE) that uses a dual-domain embedding strategy for physics-aware conditioning. Furthermore, to enable rapid inference, we develop a variance-aware acceleration mechanism that exploits learned spectral uncertainty to guide a dynamic warm-start schedule, thereby drastically compressing the sampling trajectory. Extensive evaluations on a systematically reconstructed RadioML benchmark demonstrate that FT-SDF outperforms state-of-the-art baselines. Specifically, it achieves a 59.3% reduction in sampling iterations (more than twofold inference speedup), while maintaining both high statistical fidelity (Fréchet inception distance (FID) <15) and industrial-grade precision (error vector magnitude (EVM) ≤ 14%), demonstrating the feasibility of rapid, controllable generative AI in complex electromagnetic environments.
KW - Diffusion models
KW - autonomous aerial vehicles (AAVs)
KW - edge intelligence
KW - generative AI
KW - signal generation
UR - https://www.scopus.com/pages/publications/105031944591
U2 - 10.1109/JIOT.2026.3669156
DO - 10.1109/JIOT.2026.3669156
M3 - Article
AN - SCOPUS:105031944591
SN - 2327-4662
VL - 13
SP - 23264
EP - 23277
JO - IEEE Internet of Things Journal
JF - IEEE Internet of Things Journal
IS - 11
ER -